Drug targets of other LCS compounds Our work with LCS-1 highlights the value of chemical biology for identifying new therapeutic targets and biological mechanisms in lung cancer. Will Lockwood is utilizing a similar approach to identify the molecular targets of another novel compound identified in the same screen for inhibitors of lung cancer cell lines, LCS-3. With the help of the Laboratory of Proteomics and Analytical Technology at the NCI, experiments are underway that couple affinity chromatography with quantitative proteomics to identify protein targets of LCS-3 in susceptible cell lines. Efforts have also been initiated with the NCIs Developmental Therapeutics Program to screen this compound against the NCI-60 cell line panel and in mouse models of cancer to determine whether the effects of LCS-3 extend beyond malignancies of the lung;the results may also facilitate characterization of the underlying dependencies that make tumor cells sensitive to drug treatment. Inflammation and its role in lung tumorigenesis The immune system can promote or retard cancer progression in different settings. Arun Unni has chosen to better understand this interplay by using our mutant EGFR-driven mouse model of lung cancer. In this model, we observe early recruitment of macrophages to the lung parenchyma after induction of mutant EGFR. This recruitment is well before tumor formation and suggests that signaling by mutant EGFR is capable of driving an inflammatory response. Experiments are underway to explore the mechanisms by which mutant EGFR induces expression of inflammatory genes. Studies with this and other models can help clarify the role of inflammatory cells in tumorigenesis and the mechanism by which the immune system is alerted. Secondary mutations that collaborate with EGFR The new tools of modern genomics---next generation DNA and RNA sequencing, expression arrays, SNP chips and comparative genome hybridization (CGH), and epigenetic surveys---are delivering complex information about molecular events in human tumors (cf. Ding et al. Nature 455: 1069, 2007). But these analyses are far from complete and they do not provide functional information, so there is continued need for using more traditional genetic approaches to discover the mechanisms of oncogenesis. Mutations that collaborate with mutant EGFR in lung carcinogenesis are of special interest with respect to initiation of tumors, progression to metastasis, and appearance of drug resistance. With the help of Nancy Jenkins, Neal Copeland, and their colleagues, Pang Fan (a former post-doc who has remained at MSKCC) and Will Lockwood have adapted the Sleeping Beauty transposition system to lung epithelial cells, in order to induce insertion mutations in cultured human lung cancer lines and in mice programmed to develop EGFR-induced lung cancers. For example, tumors that metastasize and tumors or cell lines that demonstrate drug resistance will be examined for sites of SB insertions to look for common insertion sites and interesting genes near the insertions. Mutations in human lung adenocarcinomas In a recently initiated collaboration with the University of British Columbia, Will Lockwood and Kreshnik Zejnullahu have begun sequencing unique cohorts of lung adenocarcinoma tumors in search of novel mutations that drive tumor development. By focusing on subsets of tumors from patients with poorly characterized mechanisms of disease initiation, they hope to enrich the sample set to favor the discovery of novel gene variants that contribute to tumorigenesis. Whole-exome sequencing data will be generated and then integrated with pre-existing SNP, CGH, DNA methylation and mRNA expression profiles from the same tumors and matched normal samples in order to uncover new genes and pathways involved in lung tumorigenesis. Functional validation of candidates will be performed by infecting cell lines and mouse models with lentiviral vectors carrying mutated versions of these genes. This approach will complement our studies of secondary mutations in mouse models of lung adenocarcinoma described above. Mutual exclusivity of KRAS and EGFR mutations In human lung adenocarcinoma, mutations in either KRAS or EGFR occur in approximately 50% of cases. However, the co-occurrence of Ras and EGFR mutations has never been convincingly observed, implying that they are mutually exclusive because of some deleterious or disadvantageous effect of co-existence in a single cell. Arun Unni has begun to test this hypothesis in vitro at both early and late stages of tumorigeneis by introducing and expressing combinations of mutant alleles in primary and established tumor cell lines. Seeking enhancers and suppressors of mutant EGFR in Drosophila. Genetic interactions are important for understanding oncogenesis and for devising novel approaches to therapy. To that end, Shenqiu Wang has prepared an expression vector to seek abnormal phenotypes that result from expression of an oncogenic form of human EGFR in various tissues of D.melanogaster. Mutations that enhance or suppress readily monitored phenotypes will be further studied to discern whether they can help identify mammalian genes that affect the function of normal and oncogenic forms of EGFR in normal cells and tumors. Role of miRNA in EGFR mutant lung cancer Many recent reports document increased and decreased levels of microRNA species in human and murine cancers of the lung and many other organs. Shenqiu Wang plans to extend these surveys to lung tumors with known classes of oncogenic mutations and then carry out functional tests by infecting mice and cell lines with viral vector libraries of miRNAs. The appearance of early tumors, metastasis, or drug resistance in infected mice or evidence for dominance of clones of cells expressing certain miRNAs would be signs of functional consequences of miRNA production and further pursued. Oncogenic susceptibility in the B cell lineage Within all developmental lineages, cells are more or less susceptible to transformation and thus have intrinsically different propensities to become a cell of origin for cancer. Our understanding of the hematopoietic system, particularly B lymphopoiesis, provides an opportunity to test the vulnerability of developmental cell stages to tumorigenesis. Arun Unni has devised a strategy, using a loxP-regulated cassette in a retroviral vector and stage-specific expression of Cre recombinase, to test whether expression of a c-Myc oncogene with concomitant loss of p53 at specific developmental points within the B cell lineage causes tumors. With these studies we will try to establish if tumors arise regardless of the stage at which an oncogenic state is created or whether certain stages of development are uniquely resistant or susceptible to transformation.